Conversation
Author
|
Ping on this one, it has been open a month with no CI run. The two lines drop a |
TobyRoseman
reviewed
Sep 30, 2026
| max_iter=300, | ||
| ).fit(weight_2d, sample_weight=importance_2d) | ||
|
|
||
| weight_2d.cpu() |
Collaborator
There was a problem hiding this comment.
Looks like weight_2d could potentially be running on CUDA, see line 666.
Author
There was a problem hiding this comment.
Yes, it can be on CUDA. Tensor.cpu() returns a CPU copy without changing the source tensor, so these two calls discard that copy and leave weight_2d and importance_2d on CUDA. The .cpu() calls on the returned centers and labels are still there. PyTorch docs.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Noticed this in the cuda branch of _cluster_weights_2d. Tensor.cpu() is not in place, so both calls build a copy and drop it while the originals stay on the device, and the assignment ten lines up shows the form that was meant. If freeing device memory before the return was the point, layerwise_compressor does it with del plus empty_cache, so you may want that here instead.